The Chitranjan S. Ranawat Award: Factors That Predict Outcome Five Years Following Total Knee Arthroplasty

Previous reports have identified several potential predictors of pain and function after total knee arthroplasty (TKA). However, the results of these studies are conflicting, and most have a short follow-up after TKA. The purpose of this study was to identify factors predictive of pain and function five years after TKA.

A multicenter cohort of 3,688 primary unilateral TKA patients from a comparative effectiveness consortium was enrolled. Demographic data, medical, and musculoskeletal comorbidities, and patient-reported outcome measures were collected preoperatively and postoperatively at five years, including the short-form health survey 36-item (SF-36) and the Knee Disability and Osteoarthritis Outcome Score (KOOS) pain and activities of daily living (ADL) function scores. Multivariate regression models with a 95% confidence interval were used to identify independent predictors of KOOS pain and function scores at five years.

The analysis identified the factors that were independently predictive for KOOS pain score at five years as follows: age, insurance, race, Charlson comorbidity index, back pain, number of other painful hip and knee joints, contralateral knee pain, preoperative SF-36 mental component summary, and baseline KOOS pain scores. For KOOS ADL, the following predictive factors were identified: age, body mass index, insurance, race, Charlson comorbidity index, back pain, number of other painful joints, contralateral knee pain, ipsilateral hip pain, preoperative SF-36 mental component summary, and physical component summary scores, and baseline KOOS ADL scores.

We have identified the factors that each independently predict less improvement in pain and function five years after TKA. These potentially modifiable factors, such as musculoskeletal comorbidities, can be targeted with preoperative patient optimization programs to improve patient outcomes and patient satisfaction after primary TKA. Knowledge of these factors that predict less improvement in pain and function can assist the surgeon and patient during shared decision-making and in setting appropriate patient expectations preoperatively.